Air Temperature and Humidity and room acoustics effect

[andonwego]

I've searched threads but they refer to the effect of temp and humidity on gear. And making studios livable.

I have a sound waves in air physics question. Speed of sound physics.


Does anybody have any direct or anecdotal experience with temp and humidity making a difference to mixing and getting masters to sound good?

I understand the degree to which this could only ever make a fringe difference.

Humidity can increase the speed of sound by .1% to .6% (Wikipedia, if anyone wants to make an issue of that as a source:

Speed of sound - Wikipedia, the free encyclopedia



At higher temperature, the speed of sound increases as well.

Usually this is an avionics thing. Supersonic Mach speed calculations and sonic booms. Air Force/Navy/etc.


But high temperature and high humidity factor or cool temp and very low humidity could, in my mind, theoretically throw off rooms with tuned acoustics and throw off things like timed millisecond delays, reverb delays to allow early reflections, and a host of other little things to change the sense and distribution of 'mud' in a mix.


This would probably mainly be an issue for classical recording, if it's an issue at all. Which I'm more than aware in most cases it isn't.

But I only ask because I don't know, and think maybe some of you might.

Thanks.

[jaddie]

Temperature changes the speed of sound significantly, humidity less so. At room temperature, a 10 degree f change makes about a 1% change in the speed of sound through air. At room temp, changing from 0% humidity to 100% humidity changes the speed of sound about .1%. However, humidity affects absorption, which is highly frequency dependent. Obviously, the more air you have to go through, the more noticeable the effect. Larger rooms will be affected more than smaller rooms.

I was working a large out-door concert once, where we had delayed speakers on towers hundreds of feet out from the main clusters. The concert was over a several days, and weather changed several times, changing temperature and humidity. We initially didn't understand why the delayed tower EQ and delay time kept changing on us. Then, after a bit of research, we attributed the delay time change to temperature change, and the EQ change to humidity change. We had a LOT of air to get sound through, so it was quite noticeable.

Temperature vs speed of sound isn't hard to find references for. You found the Wiki, here's a calculator:

Calculation speed of sound in humid air and the air pressure density of water density - sengpielaudio

However, it took some serious Googling to find references on the humidity/absobtion effect, though. It helped a lot to know it existed and was significant, but it's not talked about much. Odd, because it should be one of the more audible effects of environmental changes. I finally found this paper from Rane. This paper is about the best:

http://www.rane.com/library.html
(scroll down to "Environmental Effects of the Speed of Sound")


Jim

[andonwego]

Jim,

Do you think in a typical arena setting the effect of the crowd -- heat and humidity -- has much impact? We all know that the presence of people themselves changes the sonic absorptive quality of the floor of the arena, and front of house corrects for this, but are you aware of temperature/humidity calculations being used in an arena setting?

[edva]

Absolutely. The audience literally creates an "ocean" of humidity and heat, one effect of which is the refraction of high frequencies, a sort of "skipping" or "bouncing" towards the back of the coverage area.

[Odeon-Mastering]

Originally Posted by andonwego

Jim,

Do you think in a typical arena setting the effect of the crowd -- heat and humidity -- has much impact?

I m not Jim, but I never thought about that so far...usually you got far more importand issues to consider inside an arena (reflections being the most importand)(by the way reflections might help in easing the effect of humidity). The distances are not that far and usually delay lines are enough.
For open air gigs it is a serious issue though... one that we sort of overcome with the correct setup and use of line arrays for the main and the delayed systems . ( I remember many rainy festivals in the Uk where you had highs, for instance, here, but a step back it was completely dead)

In the studio it never even occured to me since the distance between the operator and the speakers is usually 2-4 meters.

[wado1942]

Don't forget air pressure. I've heard some producers like recording as close to sea level as possible because the rooms sound better to them.

Most acoustics environments won't be affected significantly except for larger ones. One big issue is the fact that one can get layers of different air densities in an environment. So the waves can actually bend as they travel through the air as if light through a prism. One place where this is a huge issue is in outdoor arenas. For instance, in an arena, one must factor in the time of day when most events will be taking place. The speakers need to be physically pointed above or below axis with the audience because the sound would otherwise be bent away from them.

[andonwego]

Actually that's very true. Air pressure is inevitably another factor.

Sort of the reason I brought all this up is because we have standards of audio clarity per se in the recording process, and I've recently become more interested in clinging to those standards. I've been losing a lot of my cavalier attitude toward sound, I've come to think that more perfect technique can take the place of 'art' -to a significant but not total degree- if you want to put it philosophically or whatever.

But without a sort of standard of 'purity' for playback systems, it all means less than it could.

I got the idea from looking at STP motor oil stuff. No idea if this is what that stands for, but back in high school science studying atmospheric pressure we had something called 'Measurements Made at Standard Temperature and Pressure,' or STP for short.

After reading some of the responses on here, I really believe that not only could these things probably make a significant cumulative difference, but I believe that they are probably difficult to detect. Change preamp, immediate difference. Change monitors, immediate difference. Change mic, immediate difference.

This is so very much the sort of f***** thing you'd never know to look for.


Then there's the issue of how the biological hearing mechanism changes at different temperatures, pressures, and humidity levels. The body reacts in its own ways. The microscopic structure of the ear is delicate and precise and finely tuned already as it is...

[edva]

STP - scientifically tested product. Or so I was told.

[jaddie]

Originally Posted by andonwego

Jim,

Do you think in a typical arena setting the effect of the crowd -- heat and humidity -- has much impact? We all know that the presence of people themselves changes the sonic absorptive quality of the floor of the arena, and front of house corrects for this, but are you aware of temperature/humidity calculations being used in an arena setting?

The effect of an audience in a sports-type stadium depends so much on system design and stadium design that it's hard to make generalities. Often the crowd's effect is largely acoustic, where the effect of temperature and humidity are more subtle. It's also system design dependent. If there's a big central cluster, and a lot of air for sound to go through, the atmosphere has a known and measurable impact. If the system uses a more distributed approach, the air has less impact because there's less to travel through between speaker and ear. Acoustics play a huge role here too.

I've had the unfortunate experience of trying to set up a system in a large stadium venue for a speech-centric program, where the stadium had almost no installed acoustic treatment. The room also had very high ambient noise from 4 air handlers putting noise at 75dBC SPL. In that case, air temp and humidity had an impact, but there were so many other more significant issues that it went ignored. Feedback and level control were the problems to solve.

In other venues, it's possible to anticipate ambient temperature and humidity and pre-adjust for what it will do. But the crowd often has more impact on acoustics, which again often a bigger issue.

In short, it's a factor that can be considered and compensated for, but it's far from the biggest issue to solve. If the system has significant speaker to ear distance, changes in timing and EQ can be calculated pretty well. With some of the new DSP system processors, it's possible to build a set of different calibrations that cover the range of temp/humidity anticipated. Then with sensors, trigger the changes when needed. You'd need something like 5 to 10 settings to cover the typical range, assuming properly operating HVAC, and normal head loads. Totally do-able, though I've not see it done yet.

Just give me a chance, and a client who's willing....!

Jim

"God help us. We're in the hands of engineers."

[andonwego]

Just give me a chance, and a client who's willing....!

Well yes. I think whoever has the budgetary wherewithal to execute this kind of live sound should absolutely do it. I think the generation of artists coming up, because of the reliance less on actual live musicianship and more on the subtle qualities of sound (mixers, engineers, recordists, and producers are the real artists these days for the realm of pop), would be well advised to go way beyond the Return On Investment for live sound purposes and really kill every single acoustics gremlin in sight.

Because if you're already lipsynching onstage to autotuned studio vocals anyway...

I just think, especially with your point about the availability of automated DSP programs to correct EQ and delays for these things when combined with the necessary sensors, this should all really be de rigueur. Military style. The old recording consoles had a "FIRE" button from fighter jets repurposed as the RECORD button if memory of stories is serving me correctly...

~~To the greater achievement of sonic perfection for its own sake. Hear ye.~~

[jaddie]

Originally Posted by andonwego

I just think, especially with your point about the availability of automated DSP programs to correct EQ and delays for these things when combined with the necessary sensors, this should all really be de rigueur. Military style.

Do I detect a subtle note of sarcasm?

Look, it's easy to do, and not really all that expensive, less than a few percent of total system cost. It's just that unless the space is huge, air changes make a little difference, proper EQ, alignment and acoustic design make big differences. Not every venue presents acts comprised of lip-syncing autotune artisans. There is some genuinely fine work being done today too. If a venue would invest in really first quality sound, it may be more attractive to first quality acts. That may be overly simplistic, because quality and popularity are often mutually exclusive.

A lot of shows bring their own systems anyway. So if the act cares, they could have this level of engineering.

Automatic Air Compensation (I'd call it AAC, but that means something else), must certainly have been tried somewhere. Venues I typically see are so poorly designed that it's like creating a fine sauce for an already burned goose.

My clients like me to do stuff they can clearly hear. I would probably only suggest AAC (ok, sorry, it just works out that way), for someone who has already invested in a good sounding large venue...of which there are very, very few, or an act that values the benefits of high quality sound reinforcement. Otherwise, I'd probably be wasting my time and their money.

This is a technology best applied to outdoor systems where there distances are large and air is completely uncontrolled, and adjustments like this are significant and noticeable.

Jim

[andonwego]

Yes but I used to moonlight as a writer's assistant, long story, I think/talk/type in one liners from time to time. I was being serious.

That kind of compensation should be de rigueur for everybody, always (where applicable), just another tool that *can* be used that isn't regularly (?) and would benefit everybody.

I was half kidding about the state of modern pop. Half not kidding. But I love the idea of 'AAC' per se. I'm glad I brought up this subject and thanks for sharing your knowledge.

[jaddie]

Another thought on this, one of the more sophisticated tools for sound system calibration is Audyssey. This tool works in the time domain, and has the ability to compensate for all manner of issues, acoustic and otherwise. If it were possible to run its cal routine during several sets of conditions, a really accurate automatic air comp system could be built. The system would take into account nearly every issue presented to it in one tool, though you'd probably have to have several copies of it to deal with multiple conditions. Audyssey is found in a lot of consumer devices, but there's stand-alone 8 channel pro unit too that's already found its way into a lot of large venues. Most, I believe, are well controlled environments, and mostly theaters. But there's no reason it couldn't be applied to any venue. I recommend it all the time for home theaters and studio monitors. audyssey.com for details

Just because I'm an Audyssey installer doesn't mean I'm biased. Really. But frankly, this thing is, so far, the only system that compensates for acoustic, transducer, environmental and electronic issues in one device, and does it extremely well. Probably the best tool out there to deal with the temp/humidity issue.

[Shy]

No, thanks. I find the equalization done by various "room correction" tools to have a devastating effect that can never "fix" any effects by heat, humidity or air pressure (and these change a lot in this damn desert). The main situation in which bad audible effects are caused by enviromental changes is when it's cold, with high humidity and high air pressure. This is a horrible combination, with the high air pressure providing the death blow. In this situation no type of processing can really compensate for anything and the idea that it could is unreasonable, for various technical, physical and physiological reasons.

[jaddie]

Originally Posted by Shy

No, thanks. I find the equalization done by various "room correction" tools to have a devastating effect that can never "fix" any effects by heat, humidity or air pressure (and these change a lot in this damn desert). The main situation in which bad audible effects are caused by enviromental changes is when it's cold, with high humidity and high air pressure. This is a horrible combination, with the high air pressure providing the death blow. In this situation no type of processing can really compensate for anything and the idea that it could is unreasonable, for various technical, physical and physiological reasons.

Could you share those reasons?

I completely agree that room correction tools in general are completely wrong, and have been for years. I've listened to many, always turned them off. Until Audyssey came along. It's the one exception, and unless you've heard what it does specifically, it's not fair to lump it in with all the rest.

But again, could you state for us the "various technical, physical, and physiological reasons" that make compensating for changes in air conditions are unreasonable?

[Shy]

Psychoacoustic and acoustic modeling of frequency-dependant changes caused by natural conditions are not advanced, reliable or accurate enough to not have a devastating effect in studio conditions and error margin requirements.

The largely unpredictable physiological effects of air pressure make it impossible to design an encompassing psychoacoustic model that's guaranteed to not have a negative effect for most listeners. At this point this level of psychoacoustic research is largely pseudoscientific, unproven and error-prone.

The effects of temprature, humidity, altitude, composition of the air, etc., on air absorption and the speed of sound, and hence the amplitude changes and more that are caused due to all of it, are not predictable enough to form a very accurate acoustic model on. There are unproven pseudo-scientific methods involved in designing systems that aim to predict or compensate for those changes, mostly due to the practical inability to even design a reliable testing enviroment.

Various time delay technologies are not at a level that allows artifact-free processing. Even the most advanced and well-regarded implementations of the simplest types of equalizers have undesirable effects in the most modest amounts of processing.

Due to all this and more, and my own testing of various "correction" tools (including Audyssey's), my conclusion is that these tools can't modify the sound in a way that I would consider positive (or even not negative) or a "fix".

[Odeon-Mastering]

I am not quite sure how these systems work.

In the context of live sound I think some are failing to understand that a Public Adress system is not 2 speakers which create a sweet spot in a studio. Moreover the way we mix live, is not at all, like the way we would mix in the studio
We are actually talking about multiple speakers (stacks of speakers for L/R FOH, center clusters, delayed systems, monitor world etc all of them interacting) and the main interest is to be able to have maximum headroom , great intelligibility and even distribution so everyone in the venue is listening to more or less the same thing (democracy for listeners as my fav speaker manufacturer advertises).
The response of such a system is not flat by any means .

I would be terrified if I saw some geezer with a box that automatically corrects the system, trying to plug it in after I finish setting up a large PA.

Obviously things change and the future is unknown (And I would be gladly stand corrected if someone proves to me that such a concept thing will work in action)

Anyway I think this thread is totally derailed now.

[jaddie]

Originally Posted by Shy

Psychoacoustic and acoustic modeling of frequency-dependant changes caused by natural conditions are not advanced, reliable or accurate enough to not have a devastating effect in studio conditions and error margin requirements.
<snip>

Due to all this and more, and my own testing of various "correction" tools (including Audyssey's), my conclusion is that these tools can't modify the sound in a way that I would consider positive (or even not negative) or a "fix".

Well, hmm.... the technology behind Audyssey is hardly what you describe in your first paragraph. It's not "modeling", it's measurement. It's not predicting anything, it's measuring the effects of variables. The level of resolution of its measurement is far above the standard FFT techniques. It's clear you don't understand what Audyssey is, but that's fine, I'm sure few actually do.

Please enlighten me as to the unpredictable nature of the effects of temperature and humidity. I remain confused about your statement because authoritative published data shows the effects of temperature, humidity, and altitude to a high degree of precision. In fact, if you look at the data, the effect temperature alone has is quite small. But that aside, it's not hard to accurately measure the speed of sound in a transmission medium of any kind with high accuracy. In fact, ultrasonic ranging depends on this ability. Absorption is also quite easy to quantify. Oddly, these factors don't seem unpredictable at all to me, but if you know of a study that shows otherwise, I'd be most interested. Not that it's a big deal anyway, I wasn't thinking of "predicting" air changes, as much as measuring them in-situ, and adding appropriate correction.

While you're at it, if you would please point out in specific what portion of their research could be termed "pseudoscience", I would also be enlightened. It sure seemed quite scientific and detailed to me, but I've been wrong before.

As to your test of Audyssey, I can't comment, except to say after many successful installations, I've found that there are occasions where its effect seems minimal. Those situations are marked by inherently good acoustic design and speaker performance. In other words, not much correction required, so little is provided. Try it with a more average set of conditions, though, and you begin to see the improvements. Your single, possibly cursory evaluation is a single data point. The reality requires thousands of data points to prove a trend, in this case, an audible improvement or not. You probably designed a superlative system in the first place.

The engineer in me says everything is measurable if you understand what you are measuring. If you can measure it, you may be able to correct for it. In the real world, it ain't that easy, but in time we'll get there.

Jim

[jaddie]

Originally Posted by Odeon-Mastering

I am not quite sure how these systems work.

In the context of live sound I think some are failing to understand that a Public Adress system is not 2 speakers which create a sweet spot in a studio. Moreover the way we mix live, is not at all, like the way we would mix in the studio
We are actually talking about multiple speakers (stacks of speakers for L/R FOH, center clusters, delayed systems, monitor world etc all of them interacting) and the main interest is to be able to have maximum headroom , great intelligibility and even distribution so everyone in the venue is listening to more or less the same thing (democracy for listeners as my fav speaker manufacturer advertises).
The response of such a system is not flat by any means .

Ah, excellent comment! And you might add, no concert-level FOH system goes without extensive EQ and time adjustments. None. And those that perform those adjustments know the reasons why a flat system in a large venue won't sound right, and they adjust accordingly, and probably gave up on 1/3 octave EQ long ago as useless. They know the importance of delay, and it's impact on articulation. Ideally, when it's all done, every seat is a "money seat".

Originally Posted by Odeon-Mastering

I would be terrified if I saw some geezer with a box that automatically corrects the system, trying to plug it in after I finish setting up a large PA.

Obviously things change and the future is unknown (And I would be gladly stand corrected if someone proves to me that such a concept thing will work in action)

I'll try not to take that personally! But seriously, I'm with you. I was in the same place too. But if you watched an experienced tech tune a system and heard the result and thought it was excellent, you'd feel good about hiring him again, right? The same is true for a tool of any kind. Auto-EQ has a bad reputation with good reason; it never worked, and mostly still doesn't. We've been talking about Audyssey, which I've had quite a bit of experience with. It does things automatically and quickly that used to take me hours. It also does things I could never do. That's why I'm behind it in general. But it's not just me, it's clients like IMAX theaters (whatever you think of them).

Just to put this back on the rails, I think it's a tool that if used properly could completely compensate for the effects of humidity and temperature in large venue systems. Its correction is based on extremely high resolution measurement, way above conventional 1/24th octave RTA, for example. And deals with the time-domain, which an RTA doesn't. It's got everything you'd need to air-comp a big system.

Jim

[dcollins]

Originally Posted by Odeon-Mastering

Anyway I think this thread is totally derailed now.

"For my birthday I got a humidifier and a de-humidifier... I put them in the same room and let them fight it out."
-- Steven Wright


Hope this helps!


DC